SCF^FBXL15 Regulates BMP Signaling By Directing The Degradation Of HECT-type Ubiquitin Ligase Smurf1

Ubiquitin-proteasome system plays crucial roles in many cellular processes to target specific proteins for degradation in all eukaryotes. Its dysfunction is closely associated with multiple diseases, like inflammation, cancer and neurodegenerative diseases. Protein ubiquitination is a highly ordered multi-step enzymatic cascade catalyzed by ubiquitin-activating enzymes (E1s), ubiquitin-conjugating enzymes (E2s), and ubiquitin ligases (E3s), which determine the specificity of binding substrates. According to the ubiquitin transfer mechanisms, E3s could be further classified into two categories: the RING (really interesting new gene) finger E3s and HECT (homologous to E6AP carboxyl terminus) domain E3s. SCF (Skp1-Cul1-F-box) complex is so far the best characterized RING-finger type E3 that is composed of multiple subunits including the scaffold protein Cullin1, the RING finger-containing protein Roc1, the adaptor Skp1 and the F-box proteins responsible for substrate recognition. Due to the substrate diversity and functional divergence, SCF complex plays important roles in control of cell cycle, cell growth and tumorigenesis. Compared to RING-type E3s, there still remain many more unanswered questions regarding HECT-type E3s, among which the question how HECT domain E3 activity is regulated constitutes a particularly significant gap. Human Nedd4(neural precursor cell-expressed developmentally downregulated gene 4) family with nine members is the best characterized subgroup of HECT-type E3s, and several mechanisms in multiple layers of regulation have been determined in this family, such as E3-substrate recognition and interaction, E2-E3 binding affinity and relief of intra-molecular autoinhibition.Smurf1(Smad ubiquitination regulatory factor 1) is the typical member of the HECT-type Nedd4 family ligases with a C2-WW-HECT architecture. Smurf1 was originally identified as the ubiquitination regulatory factor of Smad1/5, and provided the first link between TGF-β/BMP signaling and the ubquitination system. Genetic evidence has shown that Smurf1-deficient mice exhibit an age-dependent increase in bone mass due to augmented BMP responses. We previously demonstrated that CKIP-1 (casein kinase-2 interacting protein-1) functions as the auxiliary factor to enhance the activation of Smurf1 through direct binding to the WW domains linker. To to further investigate the potential function and regulation mechanism of HECT E3s, in this study, we performed yeast-two-hybrid screens using WW region (WW domain plus linker) and HECT domain truncates of Nedd4 family members as baits to search for new interactors in human adult liver and brain libraries. Total 180 preys were identified and one of the preys, the F-box protein FBXL15(F-box and LRR-containing protein 15) attracted us for further investigations for the possible crosstalk between two type E3s since most F-box proteins can serve as a subunit of SCF type E3 ubiquitin ligases and determine the specificity of target binding.According to the domain organization, human 69 F-box proteins fall into three major groups: those with WD40 domains (FBXWs), those with leucine-rich repeats (FBXLs) and those with other diverse domains (FBXOs). FBXL15 belongs to the FBXL subfamily and is composed of 300 amino acids involving an F-box motif in the N-terminus and six leucine rich repeats (LRRs) distributed in the resting regions. We first confirmed that FBXL15 can interact with Smurf1 in vivo and demonstrated that FBXL15 binds to Smurf1 through its C-terminal LRR region and the HECT domain of Smurf1 is responsible for FBXL15 binding, and the precise interacting region was mapped to the large subdomain of the N-lobe of HECT domain. The HECT domain C-lobe contains the catalytic cysteine and the N-lobe consists of large and small subdomains which the latter one houses the E2 binding site. Our results first showed the possible role of the large subdomain.In the aspect of regulation mechanism, we confirmed that FBXL15 serves as a subunit of SCF complex to target Smurf1 for ubiquitination and proteasome dependent degradation. Importantly, the degradation of Smurf1 by FBXL15 is dependent of the formation of an intact SCF complex and independent of Smurf1 auto-E3 activity. To our knowledge, this is the first example to show that the typical HECT-type E3 Smurf1 acts as a substrate to be degraded in trans by a RING-type E3, especially by the largest multi-subunit E3 SCF complex. In addition, Smurf1 K355 and K357 were first identified as the major ubiquitination sites for FBXL15-induced Smurf1 degradation. These findings directly extend the regulation of HECT-type E3s to the level of protein stability control other than auto-ubiquitination and provide another functional crosstalk between the two major types of E3s. We then sought to better characterize the functional relevance of FBXL15 and Smurf1 in BMP signaling. Since it has been well established that Smurf1 negatively regulates the BMP signaling pathway, and now FBXL15 negativley regulates the Smurf1 stability, we hypothesized that FBXL15 should be a positive regulator of BMP signaling. By detecting Smurf1 substrates, phosphorylated Smad1/5 level, BMP signaling activity and target gene expression, we found that FBXL15 could antagonize the effect of Smurf1 on BMP signaling. Since BMP signaling also plays crutial roles in the processes of embryonic development and adult bone remodeling and Smurf1 has been demonstrated to function in dorsal-ventral patterning of Xenopus embryos through antagonizing BMP signaling and specifically suppress bone formation in mammals, next we investigated the physiological roles of FBXL15 both in zebrafish and in mammals. Knockdown of fbxl15 expression in zebrafish embryos resulted in embryonic dorsalization with altered dorsal and ventral marker expression, mimicking bmp2b-deficient mutants which suggests an involvement of fbxl15 in dorso-ventral patterning of zebrafish embryos by regulating Bmp signaling. This is the first evidence to show a direct role for an F-box protein in the determination of the dorsalization-ventralization axis. Furthermore, we showed that FBXL15 depletion resulted in an increase of bone mass, bone mineral density, trabecular number and trabecular thickness in adult rats indicating FBXL15 is a novel bone remodeling regulator. Collectively, our data first demonstrate the role of FBXL15 in embryonic development and bone homeostasis and partially reflect the in vivo physiological relevance between FBXL15 and Smurf1.Taken together, our findings demonstrate that F-box protein FBXL15 can form an SCF complex and target HECT E3 Smurf1 for degradation. The FBXL15-Smurf1 relevance was found to be involved in the regulation of BMP signaling activity, embryonic development and adult bone homeostasis. We for the first time reveal a new regulatory mechanism that HECT type E3 can be degraded in tran by SCF complex, identify the substrates and physiological roles of FBXL15 and provide clues for further functional investigation of SCF complex.